Looks like a standard sand bar.
Maybe, maybe not. Were it not for the "bridge" area it looks like the channel would support some vigorous currents (look at how blue the water is on either side of the bay) -- which possibly argues against the natural formation of such a sandbar.
I note that 17,500 years ago was during the last ice age -- ocean levels would have been lower, so if there was a low ridgeline along that route, it might have allowed people to move on foot between the two areas. No need for a manmade bridge then, and it would explain how a sandbar might occur naturally.
Although I don't know the scale of the features in the high altitude photo, the "bridge" appears to be consistant with a naturally-formed feature. As a result of prevailing winds and currents, especially longshore currents (set up by the wave interaction with the shoreline), it is not unusual to form long sand spits and barrier bars which are sometime submerged.
Several good examples of this type of shallow water and nearshore feature are: (1) the barrier bars and beaches south of Asbury Park, NJ; (2) the Cape Hatteras area, NC; (3) Galveston Island, TX, and; (4) Morro Bay, CA.
The Morro Bay example is on a much small scale, but is a good one. Morro Rock (a craggy, volcanic hill that sits north of the bay) has been alternately connected and separated from the mainland several times over the past 400+ years. South of the Rock is a long sandy peninsula which extends south to the Montana de Oro point. The Rock today is connected to the mainland by another spit, but it is breached by heavy wave action from time to time and is then underwater. Sedimentological and archeological work I did at the Bay over 20 years ago showed that the long, linear spit has been present at least since the last ice age (about 20,000 years ago). Indian "kitchen middens" showed human inhabitation in the area for at least 6,000 years.
At one time the long linear spit may have connected to Morro Rock (today there is the small entrance to the bay and harbor, which is dredged). This seems like a reasonable analong to the situation shown in the satellite photo.
The color changes in the water are a result of both water depth and the nature of the sediments or rocks on the ocean floor. There is no way to tell how thick the sediment is beneath the ocean floor, or if it is underlain by bedrock. If the shallow sediments are underlain by bedrock, the the currents would have had to cut a channel through the isthumus (which is unlikely). Interesting picture though - I'm going to have to break out the atlas to figure out precisely where this is located.
I note that 17,500 years ago was during the last ice age -- ocean levels would have been lower, I checked a map of continent contours at -110 meters that I think blam posted a link to. Sri Lanka is a wide, Florida-like penninsula off southeast India at that sea level.
My recollection is that -110 meters was the sea level at the last glacial maximum, about 18,000 years ago.
Guys you can find more information about this thing from this site. Whatever stated is true in this article, it is true that time is not staying it has been moving on it’s own, what we know about past is very less and beyond our imagination. Hindu dharma has explained alot about past and preserved most of the things. Rmayana is a part in protecting humans from devils on earth. It is difficult to believe that once upon time there is evil on earth. How you can understand by searching differetn things. I am just giving ancient history of INDIA. Through this website. It need more than English language to understand. If interested than try your best in understanding what happened in time.
http://www.answers.com/topic/yuga